Opportunities › Android device automation and testing at scale › Proposed app
Pocketbench
Run Android tests and agents on many phones over the network, no USB or root
- Score
- 49
- confidence 42
- Evidence
- 1
- items on the page
- Rivals
- 0
- named in sources
- First version
- 60 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2008
- Mature market · AI estimate
ScreensConcept
Desktop app: Test engineers and CI systems drive the console to run scripts, manage device pools, and debug test failures from their development machines; phones are targets, not control surfaces. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
Devices
3 reserved for CI14 online
About this appAI proposal
Pocketbench lets you control Android devices over the network instead of USB cables and ADB. Install a small companion app on each phone, pair it with a code, and it joins your device pool. No root is needed and nobody has to plug anything in.
Drive devices from Python, Node.js, the command line, plain HTTP or Appium. AI agents and RPA scripts use the same session API, so a workflow that runs on one phone can run on twenty at once.
A phone-friendly dashboard shows which devices are free, what each run is doing and where it failed, so QA leads can check a night run from anywhere. Pricing follows concurrent devices, not seats.
- Tagline
- Your Android test phones, on the network instead of on your desk.
- Voice
- PracticalCalmDirect
- Why this name
- A pocket-sized phone and a test bench in one plain two-part word: a bench of phones you can reach from anywhere.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Network pairing, no cables
Pair a phone with a short code over Wi-Fi or a secure relay. Devices stay connected without USB hubs or a desk full of cables.
No member yet
No root required
The companion app works through standard Android permissions granted once, so test phones need no rooting or special builds.
No member yet
Many devices in parallel
Start the same script on a pool of devices and watch each one. Sessions are isolated and queued when devices are busy.
No member yet
One API, many clients
Use Python, Node.js, CLI, HTTP or Appium against the same sessions, so existing test suites can move over with small changes.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#1E5B4FPine · primary#F6F3ECPaper · surface#1F2A24Ink · text#E0702BSignal Orange · accent#3A7CA5Lake · data
Typography
Aa
Pocketbench
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
Your Android test phones, on the network instead of on your desk.
PricingAI proposal
The material gives no competitor prices. These tiers are a starting point to test, built on the monetization idea of pricing by concurrent devices, minutes and on-premise licensing.
Solo
Free
- 1 concurrent device
- CLI, Python and HTTP access
- Run history for 7 days
Team
$49 a month
- 10 concurrent devices
- Appium and Node.js clients
- Agent sessions with limits
- CI tokens
- Run history for 30 days
Scale
$199 a month
- 50 concurrent devices
- Device reservations by team
- Priority relay
- Run history for 90 days
What it could be worthEstimate
What a software business making this much sells for today, from real deals. Pick a monthly revenue.
If it makes
Worth about
$194,000
Sale range $132,000 to $263,000
$120,000 a year in revenue
From 87 software businesses listed and sold on Empire Flippers in the last year, revenue vetted by the broker: they sell for about 19× their monthly revenue, most between 13× and 26×.
Measured on profit, the middle deal is 32× monthly net profit: about 2.7 years of profit.
An estimate, not an offer or financial advice. A real price depends on profit, growth, churn, how much of the work is the owner's, and the buyer.
How it would growAI proposal
How the product would reach new customers through its own use, and how the first ones would be found. A plan to test, not a result.
- Built-in loop
- No loop in the product. Reaches customers through open client libraries, docs, example repos and CI templates that developers find and try. Run reports are private by default, so the product has no built-in sharing loop.
- Do they become customers?
- Developers who try the free tier can become paying teams as device counts grow. Test whether free solo users bring in teammates and convert when they need more than one concurrent device.
- First customers
- QA and mobile engineering leads at small app companies (5 to 30 engineers) who run Android tests on a handful of USB-connected phones today.
- Where to reach them
- GitHub and open examples: Publish sample Appium, Python and CI templates that run on Pocketbench, and answer issues in related open-source Android automation projects.
- Developer communities: Post worked write-ups about replacing USB cables in test setups on Android developer forums and test automation groups.
- Direct outreach: Offer a free pilot to QA leads of small app teams, setting up ten phones over a week.
- First milestone
- Five small mobile teams each pay for a Team plan after a four-week pilot of at least eight devices.
Why people already pay for this
- VerifiedGitHub repository gained 450 stars in 28 days
Tech stack
Every proposal builds on the same PWA base, installable from the browser on any phone or computer. Dashed boxes are what this app adds.
- AppNext.js and TypeScript
- PWAManifest and service worker
- UITailwind CSS
- Data and authSupabase Postgres
- PaymentsStripe
- EmailResend
- HostingVercel
- This app · Device companion appNative Android app using Accessibility and screen captureControls the phone without root or USB. A PWA cannot do this on a device.
- This app · Realtime relayWebSocket relay service with per-device tunnelsKeeps phones reachable behind NAT and streams screens and commands to clients.
RisksAI inference
- TechnicalWithout root, Accessibility-based control is slower and less complete than ADB. Some actions, system dialogs and secure screens may not be reachable, which could break test suites.
- PlatformGoogle Play and Android policies limit Accessibility use and background services. The companion app may need sideloading or a managed-device setup, and OS updates could break it.
- CommercialTeams with working ADB setups or existing device farms may see little reason to switch. Demand must be tested beyond the GitHub interest.
- TechnicalStreaming screens and commands through a relay adds latency and cost per device. Flaky Wi-Fi may cause test failures that look like app bugs.
- RegulatoryAgents driving real apps on real phones can touch personal or payment data. Customers will need clear data handling, session limits and audit logs.
Everything marked AI proposal or AI inference was written by a model from the evidence on the opportunity page. The proposed name must be checked against trademark registers before it is used.